Language impairment, or aphasia, is a common problem after left hemisphere stroke, and causes significant long-term disability. After the initial period of healing, recovery from stroke relies on plasticity in brain networks spared from direct stroke damage. Thus, to improve behavioral and biological treatments for aphasia, we must understand how spared brain structures and connections contribute to recovery. Over a century of research has demonstrated that left hemisphere areas surrounding the stroke and right hemisphere areas symmetric to the normal left hemisphere language network participate in aphasia recovery. However, the mechanisms by which these areas are recruited and their roles in language recovery remain unclear, particularly for the ?right hemisphere language network.? We propose to examine the role of spared brain structures and connections in recovery of core language functions in a large group of left hemisphere stroke survivors and matched controls. The innovative methods address major limitations of prior studies by accounting for individual differences in stroke severity, using task-independent brain structure and connectivity measures, and examining several core language functions as defined by statistical analysis of an extensive behavioral assessment battery. Preliminary studies using this new approach suggest that parts of the right hemisphere language network grow after stroke, and that these changes improve language outcomes. Based on these findings, we hypothesize that recruitment of the right hemisphere language network after left hemisphere stroke compensates for language deficits. We will test this hypothesis against multiple alternative hypotheses, using advanced brain imaging methods to test for effects throughout spared parts of the brain. Analyses will examine how spared brain structures and connections contribute to language recovery, accounting for individual differences in stroke severity. We will additionally test for signs of brain plasticity after stroke and test relationships between stroke location and patterns of brain plasticity to better understand the mechanisms of recovery. New multi- modal analysis methods will integrate the brain structure and connectivity data, providing a more complete picture of recovery of language functions than has been possible before. This research will significantly advance our understanding of the biological basis of aphasia recovery, which will be vital in order to design maximally effective treatments.